Fixed-Camera Shelf Stock Tracking with Planogram-Guided Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing stock keeping methods in retail stores using fixed cameras struggle to accurately and efficiently identify and track product units across inventory structures, especially with low overlap between camera fields of view and low product unit resolution images.

Innovation Solution

A method that involves accessing photographic images from fixed cameras, retrieving the geometry of their field of view, estimating segments of inventory structures, identifying slots, retrieving product models, extracting features from images, detecting product units, and representing their presence in a realogram, all while leveraging a store's planogram to enhance accuracy and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If fixed cameras are used for stock keeping, then automation extent is improved, but measurement precision deteriorates due to low product unit resolution images

Engineering Contradiction:

Improveautomation of stock keepingVSAvoidproduct unit identification accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The inventory structure is divided into multiple segments, each captured by a dedicated fixed camera. By segmenting the monitoring task across multiple cameras positioned at different locations, the system achieves both automation and sufficient measurement precision for each local segment without requiring high-resolution images of entire inventory structures

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system transitions from relying on single-camera high-resolution images to using multi-camera low-resolution images. By adding the dimension of spatial distribution (multiple camera positions), the system compensates for the loss of detail in individual low-resolution images through geometric reconstruction and multi-view synthesis

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Device complexity

If fixed cameras with low overlap are used, then device complexity is reduced, but reliability deteriorates due to difficulty in tracking product units across inventory structures

Engineering Contradiction:

Improvecamera system complexityVSAvoidproduct unit tracking accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The system pre-establishes geometric models of camera fields of view and inventory structure segments before actual monitoring begins. By pre-defining the spatial relationships and segmentation boundaries, the system enables reliable product unit tracking across camera boundaries without requiring complex real-time coordination between cameras

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The invention introduces an intermediary computational layer that processes images from multiple cameras with low overlap. This intermediary system uses geometric reconstruction and feature matching algorithms to bridge the gaps between camera fields of view, enabling reliable tracking across the entire inventory structure despite minimal camera overlap

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If detailed product unit identification is performed, then measurement precision is improved, but productivity deteriorates due to increased processing time

Engineering Contradiction:

Improveproduct unit identification accuracyVSAvoidstock keeping processing speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The system segments both the inventory structure and the image processing tasks. Each fixed camera processes only its local segment independently, identifying product units within its field of view. This segmentation parallelizes the processing workload, maintaining high measurement precision for each segment while significantly improving overall productivity through concurrent processing

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs partial identification actions at each camera station, focusing only on product units within each camera's specific field of view rather than attempting comprehensive identification across the entire inventory structure. This partial action approach reduces processing time per camera while the aggregation of results from multiple cameras achieves complete coverage

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20250071212A1Method for stock keeping in a store with fixed cameras

Publication Date: 2025.02.27 SIMBE ROBOTICS INC

AI summary

One variation of a method for stock keeping in a store includes: accessing an image captured by a fixed camera within the store; retrieving a field of view of the fixed camera; estimating a segment of an inventory structure in the store depicted in the image based on a projection of the field of view onto a planogram of the store; identifying a set of slots within the inventory structure segment; retrieving a product model representing a set of visual characteristics of a product type assigned to a slot, in the set of slots, by the planogram; extracting a constellation of features from the image; if the constellation of features approximates the set of visual characteristics in the product model, detecting presence of a product unit of the product type occupying the inventory structure segment; and representing presence of the product unit, occupying the inventory structure segment, in a realogram.